The human promyelocytic leukemia cell line HL-60, obtained from the American Type Culture Collection, was maintained in DMEM supplemented with 10% FBS and 100 U/ml penicillin/streptomycin at 37°C under 5% CO₂. HL-60 cells were treated with PBS or increasing concentrations of corilagin (25 and 50 µg/ml) in a 37°C incubator for different durations (0, 1, 2, 3, 4 and 5 days).

HL-60 cells were seeded in 6-well plates (1×10⁶/well) before transfection. A total of 10 µl diluted transfection reagent DharmaFECT (Tube 1) and 100 nM negative control (NC) inhibitor or miR-451 inhibitor (Tube 2) were mixed with 190 µl serum-free DMEM for 5 min at room temperature. The contents of Tubes 1 and 2 were mixed and incubated for 20 min at room temperature, then 1.6 ml antibiotic-free DMEM was added for a total volume of 2 ml transfection medium. Culture medium was replaced with transfection medium in each well. The NC inhibitor and miR-451 inhibitor were synthesized by Shanghai GenePharma Co., Ltd. Sequence of NC inhibitor was 5′-CAGUACUUUUGUGUAGUACAA-3′; sequence of miR-451 inhibitor was 5′-AACUCAGUAAUGGUAACGGUUU-3′. Following transfection at 37°C for 72 h, cells were harvested for further experiments.

Corilagin (purity ≥98%) was purchased from Shanghai Yuanye Biotechnology Co., Ltd. DMEM, FBS and penicillin/streptomycin were purchased from Gibco (Thermo Fisher Scientific, Inc.). Cell Counting Kit-8 (CCK-8) and Carboxyfluorescein Diacetate Succinimidyl Ester (CFSE) cell proliferation kits were obtained from Dongren Chemical Technology (Shanghai) Co., Ltd.. Annexin V-APC/7-AAD apoptosis detection kit was from BioLegend, Inc., BSA, RIPA lysis buffer and BCA protein quantification kit were purchased from Beyotime Institute of Biotechnology. Polyvinylidene difluoride (PVDF) membrane and ECL reagent were obtained from MilliporeSigma. Primary antibodies (all 1:1,000) against light chain (LC)3 (cat. no. 12741), autophagy-related 5 (Atg5; cat. no. 12994), pro- and cleaved caspase-3 (cat. no. 14220), Bcl-xl (cat. no. 2764) and Bax (cat. no. 5023) were obtained from Cell Signalling Technology, Inc.; p62 (cat. no. 18420-1-AP, 1:1,000), Beclin-1 (cat. no. 11306-1-AP, 1:1000), β-actin (cat. no. 20536-1-AP, 1:5,000), HMGB1 (cat. no. 10829-1-AP, 1:1,000) primary and horseradish peroxidase (HRP)-conjugated secondary antibodies (both 1:5,000; goat anti-mouse IgG, cat. no. SA00001-1; goat anti-rabbit IgG, cat. no. SA00001-2) were from ProteinTech Group, Inc. TRIzol^®, Alexa Fluor488-conjugated goat anti-rabbit secondary antibody (cat. no. A-11008; 1:500) and anti-fade mounting medium with DAPI were purchased from Invitrogen (Thermo Fisher Scientific, Inc.). DharmaFECT transfection reagent and RevertAid Reverse Transcription (RT) kit were obtained from Thermo Fisher Scientific, Inc. UltraSYBR Mixture was from CWBio.

HL-60 cells were seeded in 96-well plates at a density of 2×10⁴ cells/well and cultured overnight in a 37°C incubator. The cells were treated with PBS or increasing concentrations of corilagin (25 and 50 µg/ml) for different durations (0, 1, 2, 3, 4 and 5 days). CCK-8 was added to each well and the cells were incubated for 4–6 h. The absorbance was measured at 450 nm using a microplate reader. Cell viability was calculated as follows: Mean optical density (OD) of treated wells-mean OD of blank wells/mean OD of control wells-mean OD of blank wells.

HL-60 cells were treated with PBS or increasing concentrations of corilagin (25 and 50 µg/ml) at 37°C for 48 h, collected and washed with PBS three times. For the cell proliferation assay, cells were incubated with 10 µM CFSE for 10 min at room temperature, then heat-inactivated FBS was added to terminate the labelling reaction. The labelled cells were washed twice, resuspended in PBS and analyzed by flow cytometry (FACSCalibur; BD Biosciences). For the cell apoptosis assay, cells were stained with 5 µl Annexin V-APC and 7-AAD for 15 min at room temperature, then assessed by flow cytometry (CytoFlex; Beckman Coulter, Inc.). The percentage of early- and late-stage apoptotic cells was calculated using FlowJo v10 software (BD Biosciences).

Total RNA was extracted from cells using TRIzol according to the manufacturers' instructions. The integrity, purity and concentration of RNA were detected by Agilent 2100, Nanodrop and Qubit, respectively. After the samples passed quality control tests, the Small RNA Sample Prep kit (Illumina, Inc.) was used to produce cDNA libraries. Briefly, adapters were added to the 5′ and 3′ ends of small RNA, followed by RT-quantitative (q)PCR amplification and PAGE to separate the target DNA fragments. After the library was built, the insert size was detected using Agilent 2100, and the effective concentration (>2 nM) was quantified by qPCR. The cDNA libraries were sequenced using an Illumina HiSeq 2000 platform and 50 bp single-end reads were generated. Subsequently, the raw reads were processed and filtered for the length of 21–22 nt to obtain miRNA reads. The miRNAs were aligned with miRBase (https://www.mirbase.org) and annotated using Bowtie (http://bowtie-bio.sourceforge.net). To identify the differentially expressed miRNAs, the expression levels of miRNAs were normalized using Transcripts Per Million. Normalized expression was determined as follows: Read count of miRNA ×1,000,000)/read count of total miRNAs.

The criteria for identifying differentially expressed miRNAs were P<0.01 and log2(fold change)>1.

HL-60 cells were lysed with RIPA buffer for 30 min on ice and supernatant was collected by centrifugation at 12,000 × g for 15 min at 4°C. Protein concentrations were determined using a BCA protein quantification kit. Protein (30 µg/lane) was loaded, separated via 12% SDS-PAGE (Bio-Rad Laboratories, Inc.) and transferred onto PVDF membranes. After blocking with 3% BSA for 1 h at room temperature, the membranes were incubated with primary antibodies overnight at 4°C and with HRP-conjugated secondary antibodies for 1 h at room temperature. Finally, protein bands were exposed to ECL reagent and visualized using a chemiluminescent detector (Tanon Science and Technology Co., Ltd.). The relative protein expression was semi-quantified using ImageJ v.1.46r software (National Institutes of Health) using β-actin as the control.

Cells were fixed with 4% paraformaldehyde for 10 min at room temperature and washed three times with PBS. After blocking with 10% FBS containing 0.1% Triton for 20 min at room temperature, cells were incubated with LC3 primary antibody at 4°C overnight and subsequently stained with 488-conjugated anti-rabbit secondary antibody at room temperature for 1 h. Finally, cells were washed three times with PBS and mounted on glass slides using anti-fade mounting medium with DAPI. The stained cells were imaged using a confocal microscope (Zeiss LSM800; Zeiss AG; magnification, ×630). The number of LC3-positive puncta/cell was quantified using ImageJ v.1.46r software (National Institutes of Health).

Total RNA was extracted from HL-60 cells using TRIzol and reverse transcribed to cDNA using an RT kit according to the manufacturer's protocol. RT-qPCR was performed using an UltraSYBR Mixture in a CFX-96 Touch Real-Time PCR Detection system (Bio-Rad Laboratories, Inc.). The thermocycling conditions were as follows: 95°C for 5 min, followed by 40 cycles of 95°C for 10 sec and 60°C for 30 sec. The expression level of HMGB1 was normalized to that of the reference gene β-actin, and miR-451 to that of U6. The relative expression levels of genes were calculated using the 2^−ΔΔCq method (34). The primers sequences are shown in Table I.

The GEPIA database (http://gepia.cancer-pku.cn) was used to analyze the survival percentage of patients with AML with high or low levels of HMGB1 using 106 samples from the Genotype Tissue Expression (https://gtexportal.org) and The Cancer Genome Atlas (http://tcga-data.nci.nih.gov/tcga).

The data are presented as the mean ± SD from three independent experiments. Data were compared using an unpaired Student's t-test or one-way ANOVA followed by Bonferroni's post hoc test using GraphPad Prism 5 (GraphPad Software, Inc.). P<0.05 was considered to indicate a statistically significant difference.

The inhibitory effect of corilagin on AML cells was analyzed using CCK-8 and CFSE assay kits. The AML cell line HL-60 was treated with PBS or increasing concentrations of corilagin (25 and 50 µg/ml) for different durations (Fig. 1A). Corilagin significantly inhibited proliferation of HL-60 cells in a dose- and time-dependent manner (Fig. 1B). The cytotoxicity of corilagin on HL-60 cells was also evaluated by microscopy. The cell number was notably decreased and the cell shape became irregular following corilagin treatment for 48 h (Fig. 1C). Moreover, the fluorescence intensity of the corilagin-treated group was higher than that of the PBS group (Fig. 1D). These data indicated that corilagin inhibited the proliferation of HL-60 AML cells.

To investigate the apoptotic effect of corilagin on AML cells, Annexin V/7-AAD staining assay was used to assess apoptotic cell death by flow cytometry. HL-60 cells were treated with PBS or different concentrations of corilagin (25 and 50 µg/ml) for 48 h. The percentage of early- and late-stage apoptotic cells increased following corilagin treatment in a dose-dependent manner (Fig. 2A and B). To verify whether corilagin-induced apoptosis was triggered by the intrinsic signalling pathway, the expression levels of key proteins were detected by western blotting. Corilagin increased the protein levels of cleaved caspase-3 and Bak and decreased the level of Bcl-xl, indicating that the intrinsic mitochondrial apoptosis pathway was activated following exposure to corilagin (Fig. 2C and D).

Autophagy may function as a protective mechanism in tumor cells, and inhibition of autophagy promotes cell apoptosis (30). Microtubule-associated protein LC3 is a key autophagy-associated protein involved in autophagosome formation. Cytosolic LC3-I is converted to membrane-bound LC3-II by conjugation to phosphatidylethanolamine and localizes to the autophagosome (35). To determine the effect of corilagin on AML cell autophagy, the ratio of LC3-II/I and number of LC3-positive puncta were assessed by western blotting and immunofluorescence, respectively. Corilagin significantly decreased the conversion of LC3-I into LC3-II and formation of LC3-positive puncta. In addition, the expression levels of Atg5 and Beclin-1 decreased, while p62 expression increased following corilagin treatment. These results suggested that corilagin exhibited an inhibitory effect on AML cell autophagy (Fig. 3A-D).

Previous studies have revealed that corilagin prevents liver fibrosis by blocking the miR-21-regulated Smad signalling pathway (36,37). To assess changes in miRNA expression levels caused by corilagin, small RNA sequencing was performed. Compared with the PBS group, eight miRNAs were notably upregulated and two were downregulated in the corilagin-treated groups. In addition, the expression levels of the differentially expressed miRNAs showed a dose-dependent change when treated with corilagin (Fig. 4A). miR-451 has been reported to be abnormally downregulated in patients with AML and serves as a tumor suppressor by inducing apoptosis (25,26). The present microRNA sequencing data showed that miR-451 was notably upregulated following exposure to corilagin. RT-qPCR results verified the elevated expression of miR-451 following corilagin treatment in a dose-dependent manner (Fig. 4B).

To determine the mechanisms by which miR-451 regulates AML, target genes of miR-451 were screened. A previous study reported that the expression of HMGB1 was negatively regulated by miR-451 during cardiomyocyte A/R injury (27). The GEPIA database showed that HMGB1 levels were negatively associated with survival of patients with AML. The low HMGB1 expression group had a higher survival rate than the high HMGB1 expression group (Fig. 5A). To test whether miR-451 functions by regulating HMGB1 during corilagin treatment, the expression of HMGB1 was measured using RT-qPCR and western blotting. The mRNA and protein levels of HMGB1 were significantly decreased in response to corilagin in a dose-dependent manner (Fig. 5B-D). Knockdown of miR-451 decreased the corilagin-induced downregulation of HMGB1, indicating a negative regulation of HMGB1 by miR-451 during corilagin treatment (Fig. 5E and F).

To investigate whether regulation of miR-451/HMGB1 axis by corilagin is involved in inhibition of HL-60 cell proliferation, miR-451 was knocked down and cell proliferation was analyzed. The CCK-8 assay showed that knockdown of miR-451 promoted proliferation of HL-60 cells, indicating a tumor suppressive effect of miR-451 on AML cells. Compared with corilagin treatment-alone, knockdown of miR-451 attenuated corilagin-induced inhibition of HL-60 cell proliferation, which implied that miR-451 was involved in corilagin-induced inhibition of proliferation (Fig. 6A and B). A recent study showed that miR-451 enhances death of AML cells by targeting HMGB1 (38). As aforementioned, the present results also suggested negative regulation of HMGB1 by miR-451 during corilagin treatment (Fig. 5E and F). Therefore, it was concluded that corilagin inhibited the proliferation of HL-60 cells via the miR-451/HMGB1 axis.